Musical Instrument Sound Classification Using GoogleNet with SVM and kNN Model

Abstract

Automatic classification of musical instruments sound by machine is a challenging task in todays world. Musical Instrument Sound Classification becomes very popular in research field. A huge manual process is required to classify the musical data, to overcome this work the machine learning techniques with the pre-trained model is used in this model. The proposed system classifies the musical instruments sound using the pre-trained network model GoogleNet which is used as a feature extractor with Support Vector Machine and k-Nearest Neighbor are the two classifiers. The classification performance of SVM gives high accuracy rate when compare to kNN in this proposed work. In this system sixteen musical instruments used to find the accuracy using SVM and kNN.